Title: Validation of MRI Microstructural Biomarkers in Pancreatic Cancer Project Summary: The overall goal of this proposal is to develop, validate and evaluate novel steady- state MRI measures of tumor microvasculature (e.g. vessel size index (VSI), vascular volume fraction (VVF), and vessel density index (VDI)), which utilize long-lived intravascular magnetic nanoparticles (MNP) to study and quantify the unique microvascular characteristics of pancreatic ductal adenocarcinoma (PDAC). Such biomarkers are in high demand to better explain the poor biologic response to chemotherapy that PDAC offers secondary to its unique microenvironment. One hypothesis underlying this lack of response is impaired drug delivery secondary to the characteristic dense fibrotic stroma, which is a hallmark of PDAC. Recent evidence has further supported evidence that Sonic hedgehog (Shh), a developmental signaling pathway implicated in the development of PDAC, plays a role in the maintenance of this fibrotic stroma. Inhibition of Shh with newly developed small molecule inhibitors, have shown in very recent preclinical studies, improved vascular density, drug delivery, and biologic response. This work, however, utilized imaging modalities (e.g. DCE-MRI), which have limited applicability in the human pancreas secondary to abdominal motion, gas and their reliance on dynamic imaging. Furthermore, permeability measures (e.g. Ktrans) may not be ideal for quantifying the microvascular changes associated with targeted PDAC therapies, secondary to low blood vessel density inherent in PDAC. We have developed steady state MRI techniques using MNP and have demonstrated in preclinical PDAC xenograft models, an ability to monitor physiologic changes associated with Shh antibody. The work proposed here will extend these previous results to develop in both xenograft orthotopic and spontaneous transgenic PDAC mouse models, novel MRI derived microvascular biomarkers (VVF, VDI and VSI) utilizing MNP and validate rigorously with intravital microscopy. After validation our efforts will be aimed at comparing regions of increased VVF, VDI and VSI with chemotherapeutic delivery (18F-5FU) as measured by anatomic fusion of MRI with microPET imaging and test the hypothesis that MRI measures of VSI, VDI and VVF correlate to chemotherapeutic uptake. We will then test whether Shh inhibition demonstrates measurable changes in microvasculature, chemotherapeutic delivery and response. All MRI experiments will utilize the FDA approved MNP ferumoxytol (Feraheme (R), AMAG Pharmaceuticals). After the development and validation of these MRI methods, we will focus on the translation of this technique into humans in a pilot, proof- of-concept project aimed at validating these biomarkers with sophisticated immunohistochemical techniques in PDAC undergoing resection. Once valid, these biomarkers (e.g. VDI) will prove valuable in the evaluation of novel targeted therapies, and may be higher impact measures of indirect effects on tumor stroma than measures derived from dynamic approaches. I have the expertise, leadership and motivation necessary to successfully carry out the proposed work. During my dissertation applying novel MRI sequences to neurological disorders, I developed the desire to expand my scientific interest in MRI with a career in radiology. During my residency I developed a strong interest in abdominal imaging and treating those patients suffering from cancer. In order to gain experience in cancer research, I accepted a post-doctoral position in the laboratory of Ralph Weissleder investigating magnetic nanoparticles in cancer and inflammatory diseases. During my post-doctoral research, which was performed at a 50% effort concomitant with my clinical duties as a staff radiologist, I was productive (15 published papers (some in high impact journals (e.g. Cell, Nature Medicine)), and was successful in obtaining foundation grants aimed at investigating MRI and MNP in tumor microvasculature and as a marker of the inflammatory component associated with type I diabetes. This experience served as my foundation in cancer research, but also in revealing areas where I could improve my skills in quantitative imaging. Although my clinical experiences continue to excite me, these efforts recently have only served as the rationale for focused scientific questions aimed at improving quantitative imaging in its assessment of cancer therapeutics. The K08 Mentored Clinical Scientist Research Career Development Award will allow me to obtain knowledge and training in mathematical modeling, tumor physiology, and biostatistics and to combine these efforts with my clinical training as an abdominal and interventional radiologist. In order to achieve these goals, I have chosen a team of mentors and advisors with expertise in tumor physiology (Dr. Rakesh Jain), and quantitative MRI (Dr. A. Gregory Sorensen). The development and imaging experiments will be performed at both the MGH/Harvard/MIT Martinos Center for Biomedical imaging (Dr. Sorensen) and the Steele Laboratory (Dr. Jain), where I will take advantage of the cutting-edge state-of-the-art imaging facilities, and expertise in imaging sciences, computer sciences and quantitative mathematical modeling. This setting, in combination with a rigorous development plan including advanced coursework in signal transduction, mathematical modeling, biostatistics, and tumor physiology will be the perfect plan to realize my ambitions to be at the intersection of tumor physiology, quantitative imaging, and MRI and become an independent clinical investigator developing and evaluating quantitative MRI tools for the evaluation of novel small molecule chemotherapeutic agents in humans with pancreatic cancer.